© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 6205
    Prev
    Next
    INDEX
    Explanations

    In the field of

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron strongly activates on the very first token of each new text segment, effectively marking segment or document beginnings.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    allele
    0.65
     hypoth
    0.64
    忓
    0.63
     stear
    0.63
     GET
    0.63
     mech
    0.61
    bullets
    0.61
     ابي
    0.61
     imputed
    0.61
    𝙵
    0.61
    POSITIVE LOGITS
     với
    0.67
    หนึ่ง
    0.62
    ìm
    0.59
    ‎
    0.59
     amid
    0.58
    ısıyla
    0.57
     우리는
    0.55
     หนึ่ง
    0.55
     शुक्रवार
    0.55
     রয়েছে
    0.53
    Activations Density 0.063%

    No Known Activations